Harden WeCom pull streaming and add compatible Dify snapshot refresh configuration

.dockerignore

@@ -1,8 +1,36 @@
+# Version control
 .github
+.git
+
+# Development
 .venv
 .vscode
+.idea
+
+# Data and temp
 .data
 .temp
+*.log
+
+# Frontend build artifacts
 web/.next
 web/node_modules
 web/.env
+
+# Documentation (not needed in runtime)
+docs
+*.md
+!README.md
+
+# Tests (not needed in runtime)
+tests
+pytest.ini
+
+# CI/CD
+.gitlab-ci.yml
+.travis.yml
+
+# Misc
+*.pyc
+__pycache__
+.DS_Store

Dockerfile

@@ -1,3 +1,4 @@
+# ========== 前端构建阶段 ==========
 FROM node:22-alpine AS node
 
 WORKDIR /app
@@ -10,14 +11,20 @@ FROM python:3.12.7-slim
 
 WORKDIR /app
 
-COPY . .
-
-COPY --from=node /app/web/out ./web/out
+# 先复制依赖文件和后端运行必需源代码，尽量利用 Docker 层缓存
+COPY pyproject.toml uv.lock README.md LICENSE main.py ./
+COPY src ./src
 
+# 安装系统依赖和 Python 依赖（排除开发依赖）
 RUN apt update \
     && apt install gcc -y \
     && python -m pip install --no-cache-dir uv \
-    && uv sync \
+    && uv sync --no-dev \
+    && apt-get purge -y gcc \
+    && apt-get autoremove -y \
+    && rm -rf /var/lib/apt/lists/* \
     && touch /.dockerenv
 
-CMD [ "uv", "run", "--no-sync", "main.py" ]
+COPY --from=node /app/web/out ./web/out
+
+CMD [ "uv", "run", "--no-sync", "main.py" ]

README.md

@@ -36,6 +36,20 @@ English / [简体中文](README_CN.md) / [繁體中文](README_TW.md) / [日本
 
 LangBot is an **open-source, production-grade platform** for building AI-powered instant messaging bots. It connects Large Language Models (LLMs) to any chat platform, enabling you to create intelligent agents that can converse, execute tasks, and integrate with your existing workflows.
 
+## Enhancements in This Branch
+
+Compared with the upstream official branch, this maintained branch includes several practical enhancements focused on WeCom / WeChat operations and long-running session reliability:
+
+- **WeCom Customer Service runtime enhancements** — Redis-backed scheduling, state persistence, retry handling, token cache, and sharded processing for more stable large-volume customer service workloads.
+- **WeCom app managed customer service adapter** — Added `wecom_kf_app` support for scenarios where customer service is managed through a WeCom application authorization model.
+- **Dify conversation persistence** — Persist Dify `conversation_id` bindings in Redis and clear them on explicit session resets, reducing context loss after runtime restarts.
+- **OpenClaw WeChat adapter support** — Added OpenClaw-based WeChat integration for additional WeChat connectivity scenarios.
+- **WeCom / WeChat integration hardening** — Additional fixes and runtime improvements around session isolation, callback handling, and customer-service message processing.
+
+These enhancements are especially useful for teams running LangBot in private deployments with heavier WeCom / WeChat customer-service traffic and stronger context-continuity requirements.
+
+---
+
 ### Key Capabilities
 
 - **AI Conversations & Agents** — Multi-turn dialogues, tool calling, multi-modal support, streaming output. Built-in RAG (knowledge base) with deep integration to [Dify](https://dify.ai), [Coze](https://coze.com), [n8n](https://n8n.io), [Langflow](https://langflow.org).

docker/docker-compose.yaml

@@ -4,6 +4,21 @@ version: "3"
 
 services:
 
+  redis:
+    image: redis:7-alpine
+    container_name: langbot_redis
+    restart: on-failure
+    command: ["redis-server", "--appendonly", "yes"]
+    volumes:
+      - ./data/redis:/data
+    environment:
+      - TZ=Asia/Shanghai
+    ports:
+      - 6379:6379
+    networks:
+      - langbot_network
+
+
   langbot_plugin_runtime:
     image: rockchin/langbot:latest
     container_name: langbot_plugin_runtime